Results

 

Descriptive analysis of care homes nationally

Care home resident population

We estimate that a total of 398,000 people aged 65 years or older lived in one of 15,800 care homes in England at some point during 2016/17, and were followed for an average of 227 days during that year. This equates to an average of 274,000 people aged 65 years or older living in a care home as their main place of residence at any point in time in 2016/17. It is estimated that about 80% of the total care home population for this age group reside permanently in the homes, with the remainder of residents living in the homes temporarily.,

An estimated 193,000 older people lived in residential care homes and 213,000 in nursing homes as their main place of residence at some point in 2016/17, equating to about 135,000 residents aged 65 years or older living in residential homes and 139,000 in nursing homes at any time. Table 1 presents characteristics of the residential and nursing home population. The average age in the care homes at any point in time was broadly similar between residential and nursing homes (86 vs 85 years) and both had more female residents (74% and 69%, respectively). The levels of socio-economic deprivation, as described by the Index of Multiple Deprivation quintiles, of the areas that the residential and nursing homes were located in were more or less evenly split across the different quintiles in both care home types. Residential care homes tended to be smaller than nursing homes (38 vs 59 beds) (Table 1).

For those residents who had a hospital admission in the three years prior to either 17 April 2016 (if already resident) or to moving to a care home in 2016/17, we calculated that nursing homes had on average slightly higher levels of long-term conditions among their residents than residential care homes at any one time. Of the list of conditions included in the Elixhauser list, the majority were more prevalent in nursing homes, with the exception of hypothyroidism (13% vs 12%). Similarly, nursing homes had on average slightly higher levels of frailty among their residents, for example in levels of incontinence, mobility problems and pressure ulcers (Table 1).

Table 1: Estimated characteristics of the residential and nursing home population

 

Residential

Nursing

All residents

Average number of residents aged 65 or older living in a care home in 2016/17

135,000

139,000

274,000

Age in years

85.6 (8.0)

84.7 (8.0)

85.1 (8.0)

Men (%)

26.2

31.2

28.8

Days in 2016/17 in a care home

235 (129)

219 (132)

227 (131)

 

Residential

Nursing

All residents

Socio-economic deprivation**

     

Most deprived fifth (%)

16.9

17.1

17.0

Second most deprived fifth (%)

20.8

18.9

19.9

Middle fifth (%)

21.7

22.2

22.0

Second least deprived fifth (%)

22.4

22.1

22.2

Least deprived fifth (%)

18.2

19.7

19.0

Rural area (%)

20.3

18.3

19.3

Number of beds in the care home

38 (19)

59 (26)

48 (24)

Health conditions recorded in the three years prior to joining a care home in 2016/17††

     

Average Charlson Index

2.00 (1.62)

2.24 (1.72)

2.12 (1.68)

Average number of Elixhauser comorbidities

2.79 (2.02)

3.10 (2.15)

2.95 (2.09)

Cancer (%)

6

7

6

Cardiac arrhythmias (%)

31

33

32

Chronic pulmonary disease (%)

17

18

18

Congestive heart failure (%)

14

15

15

Deficiency anaemia (%)

8

9

8

Dementia (%)

53

54

53

Depression (%)

15

16

15

 

Residential

Nursing

All residents

Diabetes (%)

18

21

19

Fluid/electrolyte disorders (%)

23

28

26

Hemiplegia or paraplegia (%)

3

7

5

Hypertension (%)

58

58

58

Hypothyroidism (%)

13

12

12

Liver disease (%)

2

2

2

Peripheral vascular disease (%)

5

6

6

Renal disease (%)

19

20

19

Rheumatoid arthritis (%)

5

6

5

Rheumatic disease (%)

5

5

5

Valvular disease (%)

9

9

9

Average number of frailty related comorbidities

1.72 (1.39)

1.98 (1.52)

1.85 (1.46)

Cognitive impairment (delirium, dementia, senility) (%)

48

52

50

Falls or fractures (%)

48

47

47

Depression (%)

19

19

19

Functional dependency (%)

11

16

14

Incontinence (%)

9

14

12

Mobility problems (%)

18

24

21

Pressure ulcers (%)

8

13

11

Note: Numbers are calculated to provide estimates of the care home population at any one time.Numbers presented are either mean (standard deviation) or percentage.

Source: analysis by the Improvement Analytics Unit

How often do residential and nursing home residents attend A&E?

We estimate that there were in total 269,000 A&E attendances from care homes. As would be expected, care home residents attended A&E more frequently than the general population aged 65 or older (Table 2). Care home residents attended A&E on average 0.98 times per person per year, whereas overall, people aged 65 or older attended A&E 0.43 times. These figures mean that 6.5% of all A&E attendances for people aged 65 or older in England are for care home residents, even though care home residents account for just 2.8% of the older population.

There were about 151,000 A&E attendances from residential care homes and 117,000 from nursing homes. Residential care home residents had on average 1.12 A&E attendances per person per year, compared with 0.85 in nursing homes. This equates to about 32% more A&E attendances from residential than nursing homes. In residential homes, 40% of A&E attendances did not result in an admission; in nursing homes this was 35% (Table 2).

How often are residential and nursing home residents admitted to hospital as an emergency?

We estimate that care home residents aged 65 or older experienced an estimated 192,000 emergency admissions to hospital. As expected, care home residents were admitted to hospital as an emergency more frequently than the general population aged 65 or older (Table 2). Care home residents experience one of these admissions 0.70 times per year, whereas people aged 65 or older experience 0.25 of these admissions per year on average. These figures mean that 7.9% of all emergency admissions for people aged 65 or older in England are for care home residents, even though care home residents account for just 2.8% of the older population.

There were an estimated 104,000 emergency admissions for residents aged 65 or older from residential care homes and 88,000 from nursing homes. Rates of emergency admissions were higher in residential care homes than nursing homes, with residential care home residents experiencing 0.77 emergency admissions per person per year, compared with 0.63 in nursing homes.

What are residential and nursing home residents admitted to hospital for?

We estimated that residential care home residents had on average 0.30 potentially avoidable admissions per person per year; this equated to about 39% of all their emergency admissions. The most common potentially avoidable causes were pneumonia (12% of all emergency admissions), urinary tract infections (8%) and fractures and sprains (8%) (Table 2 and Figure 1).

In nursing homes, residents had on average 0.27 potentially avoidable admissions per year. This was equivalent to 43% of all their emergency admissions per person. The most common potentially avoidable causes were again pneumonia (15%), urinary tract infections (8%) and fractures and sprains (6%). There were more admissions from nursing homes for food and liquid pneumonitis than in residential care homes (5% vs 2%) (Table 2 and Figure 1).

However, these figures must be interpreted carefully. The reasons for admission to hospital for care home residents are often complex. Not all emergency admissions for these conditions will be avoidable. There may also be other conditions not in our definition of potentially avoidable admissions that were avoidable. We found that residential care home residents had on average a smaller proportion of potentially avoidable admissions than nursing home residents (39% vs 43%), but this was driven by the larger overall rates of emergency admissions in residential care homes (rather than lower rates of potentially avoidable admissions), indicating that there may be other admissions that could have been avoided.

Figure 1: Potentially avoidable emergency admissions for residential and nursing home residents aged 65 years or older in England, by reason for admission (% of all emergency admissions)

Note: Number of emergency admissions are rounded to the nearest hundred as they are estimates based on the correction applied to the data.

Source: analysis by the Improvement Analytics Unit

How long do residential and nursing home residents spend in hospital once admitted?

Our analysis shows that once admitted as an emergency, care home residents aged 65 or older on average spend 8.2 days in hospital. This is similar to the amount of time that people aged 65 or older in general spend in hospital once admitted, which is 8.4 days on average. This means that 7.7% of emergency hospital bed days are occupied by care home residents even though they only make up 2.8% of the population aged 65 or older.

There is, however, some variability between care home types, with residential care home residents spending on average 8.9 days in hospital when admitted to hospital, and nursing home residents spending on average 7.4 days.

Hospital admissions resulting in death

When care home residents are admitted to hospital as an emergency, the admission concludes in death in 12% of cases (Table 3). Given that many care home residents are nearing the end of their life and that they often prefer to die at home,, some of these residents may have benefitted from not being admitted and instead being allowed to die in the care home. The percentage of emergency admissions that resulted in death were similar between residential and nursing home residents (12% and 13% respectively).

Of the admissions for those conditions that are potentially manageable, treatable or preventable outside of a hospital setting, about 15% of admissions concluded in death in the hospital; this was similar across residential and nursing homes (15% vs 16%; Table 3). Of these admissions, those for food and liquid pneumonitis most often resulted in death in hospital, with 36% for residential care home residents and 29% for nursing home residents. Admissions for pneumonia resulted in death in hospital in 24% and 26% of cases respectively.

How do these patterns of hospital use vary by age within care homes?

Table 4 provides a breakdown of A&E attendances by care home residents by age bands (65-74, 75-84, 85+). It shows that were no substantial differences between age groups in rates of A&E attendances, percentage of A&E attendances leading to admission, rates of emergency admission and potentially avoidable emergency admissions.

The percentage of emergency admissions resulting in death slightly increased with age, with 9% of emergency admissions for residents aged 65 to 74, 11% for residents aged 75 to 84, and 13% for residents aged 85 or older. However, this is likely to reflect higher rates of death in general in the older groups.

The opposite trend can be seen in the average number of days spent in hospital, with length of stay decreasing slightly with age: residents aged 85 or older spend on average 7.8 days in hospital once admitted, compared with 8.5 days for residents aged 75 to 84 and 9.2 days for residents aged 65 to 74.

Table 2: Emergency hospital use for residential and nursing home residents and general population aged 65 or older in England

 

Care home residents (aged 65 or older)

General population aged 65 or older

 

Residential home

Nursing home

Total

Average number of people at any one time‡‡

135,000

139,000

274,000

9,751,000

Total number of A&E attendances

151,000

117,000

269,000

4,142,000

Average number of A&E attendances per person per year§§

1.12

0.85

0.98

0.43

Percentage of A&E attendances not resulting in an (emergency) admission to hospital (%)

40%

35%

38%

54%

Total number of emergency admissions

104,000

88,000

192,000

2,432,000

Average number of emergency admissions per person per year¶¶

0.77

0.63

0.70

0.25

Percentage of emergency admissions that were potentially avoidable (%)

39%

43%

41%

27%

Average number of emergency admissions per person per year that were potentially avoidable

0.30

0.27

0.29

0.07

Percentage of emergency admissions ending in death (%)

12%

13%

12%

7%

Average number of days spent in hospital once admitted as an emergency (standard deviation)

8.9 (14.5)

7.4 (12.4)

8.2 (13.6)

8.4 (14.0)

Note: Number of residents, A&E attendances and emergency admissions are rounded to the nearest thousand as they are estimates based on the correction applied to the data.

Source: analysis by the Improvement Analytics Unit

How do these patterns vary by level of socio-economic deprivation?

We grouped care homes according to the levels of socio-economic deprivation in their surrounding area. Care homes in the most deprived fifth of England were home to 47,000 older people in 2016/7 (17% of all care home residents), compared with 52,000 older people in the least deprived fifth of areas (19%). There were comparatively more residents in the middle three fifths (Table 5).

Rates of emergency admissions increased with increased levels of deprivation; for example, care home residents in the least deprived areas had on average 0.64 emergency admissions per person per year, compared with 0.81 per person per year in the most deprived fifth of England (Table 5).

Similarly, the rates of A&E attendances increased with increased levels of deprivation; for example, care home residents in the least deprived areas had on average 0.84 A&E attendances per person per year, compared with 1.19 per person per year in the most deprived fifth of England (Table 5). Conversely, the percentage of A&E attendances that resulted in emergency admission decreased with increased levels of deprivation: in the least deprived areas, 65% of A&E attendances resulted in admission, compared with 59% of A&E attendances in the most deprived areas. This indicated that deprived areas may have more inappropriate A&E attendances than less deprived areas.

Rates of potentially avoidable emergency admissions showed a similar trend, with slightly lower rates for care home residents in less deprived areas (0.26 vs 0.34 in the most deprived areas). However, the proportion of emergency admissions that were potentially avoidable were more similar between areas (40% in the least deprived areas compared with 42% in the most deprived areas).

The average number of hospital bed days following an emergency admission ranged from 8.1 days in the least deprived areas to 8.4 in the most deprived areas.

These findings may suggest that people living in care homes in the poorest parts of England may receive lower quality care. However, the proportion of the emergency admissions that are potentially avoidable was similar at 42% in the most deprived areas, compared with 40% in the least deprived areas. Furthermore, we found that about 47% of the care home population in the most deprived areas had a diagnosis of dementia recorded in the previous three years, compared with 39% in the least deprived areas (results not shown). Similarly, 18% of care home residents in the most deprived areas had renal disease, compared with 15% in the least deprived areas (results not shown). Therefore, some of the disparity in emergency admissions may be due to higher levels of ill health in the care homes in the most deprived areas. This could be linked to differences in, for example, the availability of publicly funded care home beds and home-based support.

Table 3: Potentially avoidable emergency admissions that resulted in death in hospital for residential and nursing home residents aged 65 years or older in England, by reason for admission (% of emergency admissions for that condition)

 

% of admissions resulting in death in hospital

 

Type of care home resident

 

Residential

Nursing

All

All emergency admissions

12%

13%

12%

‘Potentially avoidable’ conditions

15%

16%

15%

Pneumonia

24%

26%

25%

Urinary tract infections (UTI)

8%

8%

8%

Fractures and sprains

9%

6%

8%

Acute lower respiratory tract infections

8%

8%

8%

Food and liquid pneumonitis

36%

29%

31%

Chronic lower respiratory tract infections

9%

11%

10%

Intestinal infections

8%

8%

8%

Diabetes

8%

8%

8%

Food and drink issues

7%

6%

6%

Pressure sores

15%

12%

13%

Source: analysis by the Improvement Analytics Unit

Table 4: Emergency hospital use by residential and nursing home residents aged 65 or older in England, by age band

 

Care home residents

 

Age 65 to 74

Age 75 to 84

Age 85 or older

Total (aged 65 or older)

Average number of people at any one time***

32,000

82,000

160,000

274,000

Total number of A&E attendances

32,000

83,000

154,000

269,000

Average number of A&E attendances per person per year†††

0.97

1.02

0.97

0.98

Percentage of A&E attendances resulting in an (emergency) admission to hospital (%)

61%

62%

62%

62%

Total number of emergency admissions

23,000

59,000

110,000

192,000

Average number of emergency admissions per person per year‡‡‡

0.70

0.73

0.69

0.70

Percentage of emergency admissions that were potentially avoidable (%)

40%

41%

41%

41%

Average number of emergency admissions per person per year that were potentially avoidable

0.28

0.30

0.28

0.29

Percentage of emergency admissions ending in death (%)

9%

11%

13%

12%

Average number of days spent in hospital once admitted as an emergency (standard deviation)

9.2 (18.2)

8.5 (14.3)

7.8 (12.0)

8.2 (13.6)

Note: Number of residents, A&E attendances and emergency admissions are rounded to the nearest thousand as they are estimates based on the correction applied to the data.

Source: analysis by the Improvement Analytics Unit

Table 5: Emergency hospital use by residential and nursing home residents aged 65 or older, by level of socio-economic deprivation of the local area

 

Care home residents aged 65 or older

 

Most deprived fifth

Second most deprived fifth

Middle fifth

Second least deprived fifth

Least deprived fifth

All care home residents

Average number of people at any one time§§§

47,000 (17%)

54,000 (20%)

60,000 (22%)

61,000 (22%)

52,000 (19%)

274,000

Number of care homes in this group

2,900

3,400

3,700

3,200

2,500

15,800

Total number of A&E attendances

55,000

59,000

56,000

55,000

44,000

269,000

Average number of A&E attendances per person per year¶¶¶

1.19

1.09

0.93

0.90

0.84

0.98

Percentage of A&E attendances resulting in an (emergency) admission to hospital (%)

59%

61%

63%

63%

65%

62%

Total number of emergency admissions

37,000

41,000

40,000

40,000

33,000

192,000

Average number of emergency admissions per person per year****

0.81

0.76

0.67

0.65

0.64

0.70

Number of emergency admissions that were potentially avoidable (%)

16,000 (42%)

17,000 (42%)

17,000 (41%)

16,000 (41%)

13,000 (40%)

79,000 (41%)

Average number of emergency admissions that were potentially avoidable attendances per person per year

0.34

0.32

0.27

0.27

0.26

0.29

Average number of days spent in hospital once admitted as an emergency (standard deviation)

8.4 (14.2)

8.3 (14.1)

8.0 (13.4)

8.2 (13.1)

8.1 (13.3)

8.2 (13.6)

Note: Number of residents, A&E attendances and emergency admissions are rounded to the nearest thousand and number of care homes to the nearest hundred as they are estimates based on the correction applied to the data.

Source: analysis by the Improvement Analytics Unit

Review of four case studies

The IAU has to date evaluated four initiatives to improve health and care in care homes: Principia enhanced support in Rushcliffe, enhanced support for Sutton Homes of Care, Wakefield Enhanced Health in Care Homes and Nottingham City enhanced package. A summary of the findings from the IAU studies of the four care home interventions related to A&E attendances, emergency admissions and potentially avoidable admissions are included in Box 2; for a full description of the intervention and evaluation please refer to the individual reports,,,,.

For several of the IAU studies we concluded there were reductions in at least some measures of emergency hospital use for residents who received the enhanced support, compared with their comparison groups: in Rushcliffe we found care home residents were admitted to hospital as an emergency 23% less often and had 29% fewer A&E attendances; Nottingham City care home residents had 18% fewer emergency admissions and 27% fewer potentially avoidable admissions; and in Wakefield residents had 27% fewer potentially avoidable emergency admissions. In Sutton, however, the results were inconclusive across all three measures of emergency hospital use.

There were many common themes across the four initiatives, however each site differed in what they implemented and how. By comparing the four IAU studies we aimed to identify themes that may point towards the successful implementation of the improvement programmes. Although it was not possible to unequivocally identify any elements of the improvement programmes that were particularly important for a successful intervention, we did identify some elements that may be driving the results.

Access to additional clinical input

What Rushcliffe and Nottingham City (the sites that had lower rates of overall emergency admissions compared with their control groups) had in common was the aim of having an aligned general practice for each care home and within each one a consistent named GP who regularly visited the care home (weekly, fortnightly or monthly). However this was not achieved in Nottingham City. Principia estimated that about 90% of residents were registered with the aligned GP in Rushcliffe. In Nottingham City, we estimated that about 79% of residential care home residents and 76% of nursing home residents in the study had an aligned GP (in other words were registered with the most common general practice within that care home) during the period of the study. Nottingham City CCG discontinued the GP Local Enhanced Service for Care Homes in June 2018.  

In Wakefield, where the IAU study found lower rates of potentially avoidable emergency admissions but no difference in overall emergency admissions, the vanguard set out to implement GP alignment, however this was not achieved during the period of the study. Additional clinical input was provided through multidisciplinary teams (MDTs) comprising of professionals from areas including mental health, physiotherapy and nursing. The MDTs used a screening process to identify care needs that could lead to inappropriate emergency hospital use if not addressed, which may explain why the vanguard residents had fewer potentially avoidable emergency admissions than the matched control group.

Partnership and co-production

The level of co-production – in terms of joint working between health and social care providers and commissioners – may also have been an important aspect.

Nottingham City Clinical Commissioning Group (CCG) first introduced initiatives to improve health outcomes among care home residents in 2007 and these have steadily grown over time and reportedly resulted in more active interaction between commissioners and health services. Nottingham City CCG and Nottingham City Council have a joint contract in place with care homes, with a shared service specification and quality framework, with work in this area first put in place in 2011.

In Rushcliffe, the enhanced support was developed by Principia, a local partnership of general practitioners, patients and community services, together with care home managers and the Rushcliffe CCG lead. The Principia enhanced support in Rushcliffe had a long-standing programme of work to build relationships across organisational boundaries, engaging care home teams. It is possible that this has led to greater mutual understanding of the nature of the problems that need to be addressed and, therefore, more effective interventions. There was a strong focus on building good relationships and partnership working between the members of the team. Furthermore, there were regular meetings between care providers working in different settings, which promoted shared ownership and a consistent approach. These included a monthly task group meeting between representatives of all members of the team and a bi-monthly care home managers’ network, facilitated by Age UK.

Although we did not see any conclusive evidence of the effect on emergency hospital use in Sutton, it also had partnership working. In May 2014, it established a Joint Intelligence Group, consisting of representatives from all partners across the health sector with a statutory responsibility for care homes, which met monthly to share intelligence across health and social care and promote quality assurance and safety.

Difference in effect in residential and nursing homes

Both the Rushcliffe and Nottingham City studies found that the positive results were driven by significantly lower rates of emergency hospital use in the residential care homes receiving the intervention, compared with their matched control groups, while there was no conclusive evidence in either site’s nursing home subgroup.

In Sutton, the data was in general inconclusive, which is not surprising given the small sample sizes within the subgroups (77 intervention and 76 matched control residents in the residential care home group and 220 in the intervention and matched control groups in nursing homes). In Wakefield, it was not possible to do a subgroup analysis on care home type as 84% of residents in the study lived in nursing homes. It may be that the higher proportion of nursing home residents in Wakefield (84%) and Sutton (74%) made it more difficult to see an effect (compared with 48% in Nottingham City and 64% in Rushcliffe).

However, the promising results in potentially avoidable admissions in Wakefield show that it may still be possible to improve care in nursing homes, but that potentially it requires care that is more targeted to individual residents’ needs, for example through the MDTs.

Maturity of the intervention and length of the study period

Both the Rushcliffe and Nottingham City set of interventions had developed and had time to mature by the time the study started. In Wakefield, the evaluation started from the beginning of the implementation. In Sutton, not all parts were implemented by the start of the study. This is likely to have impacted on our ability to find a significant effect of the intervention in Wakefield and in particular Sutton.

Furthermore, in Rushcliffe and Nottingham City, the study period was notably longer than in Sutton and Wakefield. The outcomes in Rushcliffe were analysed over a period of 23 months and in Nottingham City the study period was 2 years and 7 months, compared with Sutton and Wakefield where the study period was 15 and 13 months respectively. This will have allowed more time for the intervention to impact on residents’ use of health care and for the study to identify a significant change.

A subgroup analysis in Wakefield, only including residents who were in the study for at least 3 months, showed a larger reduction in potentially avoidable admissions compared with the main analysis, indicating that the impact of the intervention may differ as the intervention matures and as residents have more opportunity to be impacted by the changes in the care provided.

Training

There was a strong element of care home staff training, in particular in Rushcliffe and Wakefield, where training was delivered by either community nurses or the MDT on many topics, including falls prevention and pressure sore prevention. In Nottingham City, the Dementia Outreach Team provided staff training focusing on dementia. In Sutton, tailored e-training on continence care, dementia care and person-centred thinking was delivered to care home staff, as well as some ad-hoc face-to-face training delivered by the multidisciplinary care home team when it identified a need. However there were some challenges with training implementation.

Box 2: Findings from the IAU about the impact of initiatives to reduce emergency hospital use by care homes

Principia enhanced support for care home residents in Rushcliffe,

The Principia enhanced support included a series of initiatives to improve care for people in care homes in Rushcliffe, Nottinghamshire. The initiative included aligning care homes with general practices, regular visits from a named GP, improved support from community nurses, independent advocacy and support from the third sector, and a programme of work to engage and support care home managers.

Our analysis, based on the period August 2014 to August 2016, found that care home residents who received enhanced support had 23% fewer emergency admissions (95% confidence interval (CI) 39% to 3% fewer) and 29% fewer A&E attendances (95% CI 43% to 11% fewer) than a matched control group. Potentially avoidable admissions may also be lower but the results did not reach statistical significance (28% fewer, 95% CI 0% to 49% fewer).

When we looked at residential and nursing homes separately, residential care home residents had 40% fewer emergency admissions (95% CI 58% to 14% fewer), 50% fewer potentially avoidable admissions (95% CI 70% to 18% fewer) and 43% fewer A&E attendances (95% CI 60% to 19% fewer) than the matched control group. In nursing homes, there was no conclusive evidence of a difference in outcomes.

Enhanced support for Sutton Homes of Care residents

The Sutton Homes of Care Vanguard in south London worked to improve the quality of care offered to care home residents through a range of initiatives. These included measures to improve integrated care, such as the hospital transfer pathway (the ‘Red Bag’ scheme) and weekly health and wellbeing reviews for residents. The vanguard also supported ongoing education and development for care home staff, and promoted quality assurance and safety, for example through a joint intelligence group to share information among local health and care partners and a dashboard to benchmark care home performance.

Overall, the IAU found no conclusive evidence that people moving to a Sutton care home between January 2016 and April 2017 had different emergency hospital use to a matched control group. However, there were some indications that the Sutton residents may have experienced more emergency admissions than the matched control group, including more potentially avoidable admissions. There were some indications that Sutton residents who moved to a care home during the second eight months of the study period experienced fewer visits to A&E departments than the corresponding control group, although this did not seem to result in fewer admissions.

A subgroup analysis of residential and nursing homes showed mostly inconclusive results, apart from potentially avoidable admissions in nursing homes, where Sutton residents were estimated to have 122% higher rates of such admissions (95% CI 19% to 327% higher) than the matched control group.

Wakefield Enhanced Health in Care Homes

The enhanced support commissioned by Wakefield Enhanced Health in Care Homes had three main strands: voluntary sector engagement, a multidisciplinary team (MDT) and enhanced primary care support. It aimed to improve coordination of care, reduce unnecessary hospital admissions, shorten time spent in hospital, reduce isolation, and make sure that all residents had an end-of-life care plan with a preferred place of dying. In particular, the MDT used a screening process to identify care needs that could lead to inappropriate emergency hospital use if not addressed. The IAU evaluated the first phase of an enhanced support initiative for older people living in 15 nursing and residential care homes in Wakefield between February 2016 and March 2017.

During this period, we found that residents receiving the enhanced support had 27% fewer admissions for potentially avoidable conditions than the matched control group (95% CI 45% to 2% fewer). However, there was no conclusive evidence that there was an effect on A&E attendances or emergency admissions.

No subgroup analysis by residential and nursing home was carried out, as the majority (84%) of care home residents in the study lived in nursing homes.

Nottingham City enhanced package

Nottingham City Clinical Commissioning Group commissions an enhanced package of care for its care home residents aiming to improve health outcomes and the care delivered. Elements of this package were first introduced in 2007 and have been developed further since then. In September 2014, this package of care consisted of the Dementia Outreach Team, the Care Home Nursing Team, Age UK Nottingham and Nottinghamshire advocacy, and GP Local Enhanced Support. The IAU evaluated the effects of the enhanced package of care for care home residents between September 2014 and April 2017.

We found that Nottingham City residents had 18% fewer emergency admissions (95% CI 30% to 5% fewer) and 27% fewer potentially avoidable emergency admissions (95% CI 41% to 11% fewer) than a matched control group. There was no discernable difference in A&E attendances overall, but there were 20% fewer A&E attendances that did not result in admission (95% CI 34% to 3% fewer).

When looking at residential and nursing homes separately, we found that residential care home residents in Nottingham City had 34% fewer emergency admissions (95% CI 47% to 18% fewer), 39% fewer potentially avoidable emergency admissions (95% CI 54% to 18% fewer), 20% fewer A&E attendances (95% CI 33% to 4% fewer), and 35% fewer A&E attendances that did not result in admission (95% CI 50% to 16% fewer). We found no difference in hospital use between nursing care home residents in Nottingham City and the control group.

Source: analysis by the Improvement Analytics Unit

8,000 residents lived in both a residential and a nursing care home in 2016/17.

** Socio-economic deprivation quintile is estimated based on the Index of Multiple Deprivation (IMD) 2015, available at Lower Super Output Area (LSOA) level. The LSOA is derived from the postcode of the care home where the patient is residing.

†† Health conditions, including the Charlson Index and Elixhauser list of comorbidities, are calculated on the subset of residents who had a hospital admission (emergency or elective) in the three years prior to the study. We estimated that 22% of residents at any time did not have a hospital admission during that period.

‡‡ The number of people living in care homes varies from month to month. These figures are averages across all months in the 2016/17 year.

§§ Allows for not all individuals being in the care home for the entirety of the 2016/17 year.

¶¶ Allows for not all individuals being in the care home for the entirety of the 2016/17 year.

*** The number of people living in care homes varies from month to month. These figures are averages across all months in the 2016/17 year.

††† Allows for not all individuals being in the care home for the entirety of the 2016/17 year.

‡‡‡ Allows for not all individuals being in the care home for the entirety of the 2016/17 year.

§§§ The number of people living in care homes varies from month to month. These figures are averages across all months in the 2016/17 year.

¶¶¶ Allows for not all individuals being in the care home for the entirety of the 2016/17 year.

**** Allows for not all individuals being in the care home for the entirety of the 2016/17 year.

Previous Next